ABSTRACT Theoretical models of neural circuits aim to provide a set of fundamental principles that capture neural dynamics and explain neural computation. While their value scales with the range of dynamics they can capture and explain, the models? validity depends on their ability to make accurate predictions in the face of defined perturbations. The goal of this specific research project is to employ defined perturbations that definitively test core predictions of the new models. Unquestionably, the most powerful approach for perturbing neural activity is optogenetics, yet one photon optogenetics is not up to the task. Therefore, we will use a novel all-optical multiphoton approach that combines calcium imaging with multiphoton holographic optogenetics. This new technology allows large ensembles of neurons to be targeted with single cell resolution for photo-activation in three dimensions. Using this new approach, we will test a concerted set of concrete predictions with extremely well-defined perturbations, that were never before possible. These optogenetic perturbations thus close the loop of iterative model refinement and testing to arrive at the most powerful and predictive model yet of any cortical circuit.